Class on the topic of mathematical machine learning in the OvGU Magdeburg (2017-2018, winter term):
This class will be on the theme of mathematical machine learning. Maurilio Gutzeit and Joseph Lam will respectively teach the exercise session and the tutorial. More information here and on the OvGU e-learning platform.
Class on the topic of mathematical machine learning (Statistische Lerntheorie) in the Humboldt University and in Potsdam University (2016-1017):
This class is organised in the Humboldt University and in Potsdam University by Prof. Dr. Gilles Blanchard (Universität Potsdam), Dr. Jana de Wiljes (Universität Potsdam), Dr. Martin Wahl (Humboldt-Universität zu Berlin) and myself.
Here is the material for the fourth part of the class (bandit theory): class 1 and first exercise sheet, class 2, class 3, class 4 (the last part of the class notes is required for the exam). Some comments about the exam Comments.
Seminar on the topic of statistical learning theory (Statistische Lerntheorie) in the Humboldt University and Potsdam University (2015-1016):
This seminar is organised in the Humboldt University and in Potsdam University by Prof. Dr. Gilles Blanchard (Universität Potsdam), Dr. Jana de Wiljes (Universität Potsdam), Prof. Dr. Markus Reiß (Humboldt-Universität zu Berlin) and myself. Here is a short description of the topics.
Initial meeting : Tuesday, October 13th, 2015, at 13:15. Venue is at Rudower Chaussee 25, Raum 1.114 (HU Mathematik).
Seminar format : The seminar will take the form of a block seminar, a prior to be held on Feb. 24-26 2016.
Here is the seminar material, with the topics and the assignations to the participants (to be finalised). The numerotation of topics has been readjusted with respect to what was handed out in the initial meeting.
Monte Carlo methods are concerned with the use of stochastic simulation techniques for statistical inference. These have had an enormous impact on statistical practice, especially Bayesian computation, over the last 20 years, due to the advent of modern computing architectures and programming languages. This course covers the theory underlying some of these methods and discuss how they can be implemented and applied in practice. It is mainly a methodological class.
The following topics will be covered: Techniques of random variable generation, the plug-in principle and its applications to integration, Bootstrap, Markov chain Monte-Carlo (MCMC) methods for Bayesian inference.
There will be 8 lectures on this topic on Mondays and Wednesdays at 10.
Here are the Lecture notes for the second half of the part III class "Time Series and Monte-Carlo Inference" for the Lent term 2014. These lecture notes might get modified (last update: 4 March).
The exemple class session will be held on Wednesday, the 12th of march, at 4-5:30pm. Here is the exemple sheet.